The Local Optimality of Reinforcement Learning by Value Gradients, and its Relationship to Policy Gradient Learning
Michael FairbankEduardo AlonsoPublished in: CoRR (2011)
Keyphrases
- reinforcement learning
- policy gradient
- actor critic
- learning process
- learning algorithm
- reinforcement learning algorithms
- policy search
- function approximation
- model free reinforcement learning
- policy gradient methods
- state action
- function approximators
- action selection
- supervised learning
- learning problems
- learning tasks
- temporal difference
- optimal control
- temporal difference learning
- reinforcement learning methods
- sample size
- optimal policy
- search space
- machine learning